AJE AI scripts

Posted: August 25, 2012 in News bout Ageod games, Wargaming

As I know some beta of AJE are sometimes coming here, Imho the followingAI command is wrong

AI.SetAggro  = $Theater_Hispania;-95;$Theater_Gallia;-95;$Theater_Italia;150;$Theater_Illyria;50;$Theater_Macedonia_Graecia;50;$Theater_Thracia;100;$Theater_Asia_Minor;750

AI.Setggro doesn’t allow negative values ( unless things have changed of course). It would rather be AI.ChgAggro.

For those interested, there are AI events files. It’s a little more developed than in PON, and should produce a not fully dumb AI. I believe my own way produces a much less scripted AI, now of course, I may be wrong. Only play will tell the truth. To have played with this sort of scripting for SVF  3 years ago, it gives some results without reaching the current state of AI in Fatal Years.Let’s hope I’m wrong!

  1. Bohémond says:

    Tahnk you for spotting this. It will be fixed.

    AI in AJE is still WIP. And I think improving IA is a never ending task. So we will try to improve it event after release.


    • Clovis says:

      I don’t have any doubt about; The key is to learn how the AI “reads” the map,forget to try to force AI to mimic human play and understand design must be developped in the very same time than AI, not at the end. I’m going to seeif it’s possible to boost Spartacus AI without cheat to win against Human player.


  2. Wargamer says:

    Clovis, I am sure you are the prodigal son 😉

    • Clovis says:

      Absolutely not. I’m living well in my buddhist monastery, occupying myself to my Fatal years bonzaï. It’s just I’m always eager to discover new way of drive the AI. I’m eager to try AJE as they have chosen to put very higher value than I for Theaterprevalence and AI.LocalInterest commands; i wonder how the AI will act, and if some positive lessons may be implemented, I will. Personnaly, I never got beyond 800 ( For Baku and Manassas) and never found time to test above this value.

      the Spartacus scenario starts with a mandatory move for Spartacus stack, before a north/South choice. Aggro isn’t randomized ( in FY it is, and it works, never design AI to mimic human, just use the tools to create AI versatility and surprise), some cities will become the objecytives for AI, with possibly a great predictability of AI moves after one or two games). As it’s a short scenario in scope, it should give some results ( at least AI will not run to try to capture Lugdunum :-)). For great campaigns, it could be unsufficient. Now, that’s only guesses from my experience ( it’s really close of my SVF 1.0 AI, minus theaterprevalence tweaks as the command wasn’t yet existing) and the current AJE Hard AI may well have progressed to the point to need only this sort of AI events files. Or these AI events with these values may be better than my work too…That’s why I will play it at the end of September, to draw the lessons for my bonzais.There’s a spanish review where the Slave AIwas beaten by the reviewer after having rightly targeted harbours to leave for Sicilia. Rather good AI behaviour, even if I feel it missed the spot early against an unexperienced human believing in the invinvibility of Roman troops.

  3. Wargamer says:

    IIRC the prodigal son was not precisely in a buddhist monastery 🙂

    Time will tell… 😉

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s